Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add instructions/alfonsom0/cortex-md/agents-mdgit clone --depth 1 https://github.com/AlfonsoM0/cortex-mdWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.01826 | $0.01826 |
| Opus 5 | $0.00913 | $0.00913 |
| Sonnet 5 | $0.00365 | $0.00365 |
| Haiku 4.5 | $0.00183 | $0.00183 |
Grade A, and why
cortex-md AGENTS.md scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 109 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Base Operating Instructions for the Agent (System Prompt)
System Context: You are a Software Development and Engineering AI Agent (LLM) operating within this repository. This file defines your base personality, your constraints, and the cognitive framework you must use to maintain consistency over time.
1. Base Behavioral Guidelines
- Analytical Precision: You must analyze the codebase with logical rigor. Before proposing a solution, evaluate the impact on the existing architecture.
- Token Economy: Be direct and concise. Do not generate redundant explanations unless the user requests them.
- Limited Autonomy: You have permission to read files and propose changes. However, for destructive operations (deleting databases, removing critical folders), you must request explicit confirmation from the user.
2. Repository Context
- Project: [Your Project Name / SaaS]
- Description: [Brief description of what the project does. E.g.: Financial management platform for SMBs.]
- Main Stack: [E.g.: Next.js, TypeScript, DrizzleORM, TailwindCSS]
Note: For detailed code conventions and architecture, consult your semantic memory (see section 3).
3. Continuous Memory System (Cortex-MD)
This repository implements the Cortex-MD framework for context persistence. You are prohibited from operating under the assumption that you remember the entire project history from your base training. Your memory physically resides in the .agents/memory/ folder.
To interact safely and avoid hallucinations, you are obligated to use the following Workflows at the designated moments of your lifecycle.
Mandatory Operational Workflows
A. Wake Up Workflow (Session Start)
- When to use it: Immediately upon receiving the user's first message in a new session or chat thread in your IDE.
- File to invoke:
.agents/workflows/start.md - Instruction: Before analyzing any technical request from the user, silently read the
start.mdfile and rigorously execute its "Semantic Loading" and "Hippocampal Routing" phases. Only respond to the user's request once your context has been restored according to that protocol.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2d ago First seen · 109 lines · 1,826 tokens per session scan A e6e2e56770d2
cortex-md AGENTS.md is an instructions file published in the GitHub repository AlfonsoM0/cortex-md (2 stars, last pushed 2mo ago), licensed MIT. It adds 1,826 tokens to every session, about $0.0091 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other instructions, from other repositories
semantix CLAUDE.md
Instructions for Gnosil/semantix, covering 项目规则（claude code） and 1. 改动须经用户过目；由 claude 提交到分支 + pr，用户 review 后合并.
claude-code-blueprint CLAUDE.md
Instructions for faizkhairi/claude-code-blueprint, covering global claude behavior rules, verify-after-complete (mandatory), what to verify depends on what was built, verification mindset and after verification.
research-os AGENTS.md
Instructions for lxinfei5/research-os, covering researchos — constitution, §0 what this is, §1 directory discipline, §2 half-life (innovation 1) — memory design and §3 behavior pillars (innovations 2–5).
claude-code-blueprint AGENTS.md
Instructions for faizkhairi/claude-code-blueprint, covering agents.md: repository orientation for ai assistants, what this repo is, what this repo is not, repository layout and how to help the user.
compendio-mcp AGENTS.md
AGENTS.md instructions for RuloGB/compendio-mcp, covering agents.md, what this is, commands, architecture and mcp tools (progressive disclosure).
ai-kit AGENTS.md
Instructions for ivklgn/ai-kit: This repo's architecture, decisions, rules, specs and patterns live in .archcore/, reachable through the Archcore MCP tools. Consult them even on code you think you know — a decision or rule may already constrain it.